Buying Breakouts: Why New All-Time Highs Beat Dips

"Buy low, sell high." It's the most repeated piece of investment advice ever written. And, for the most part, it's wrong.

In 2013, the Nobel Prize in Economics went to three economists — Eugene Fama, Lars Peter Hansen, and Robert Shiller — for their empirical analysis of asset prices. One of their central findings sums up, better than any trading manual, how markets actually work: in the short run, prices are essentially impossible to predict; but over longer horizons, of several years, trends do emerge and they can be anticipated.

That single sentence captures an entire trading philosophy. And it's the foundation of what we do at ATH Scanner.

The short term is noise. It's not where to play.

In the very short term — minutes, hours, days — markets are noise and randomness. Fama showed this sixty years ago: new information gets priced in almost instantly, and what's left is an erratic walk that no one can consistently predict.

There's also a practical reason to stay away. In that arena you're competing against the world's best algorithmic trading firms — infrastructure, data, and speed that no individual operator can match. It's a game that's lost before it starts.

Step back, though. Drop the 5-minute candle, look at the monthly chart, and the noise fades. What emerges is something else: the trend. That's the timeframe where a systematic trader can actually compete — not by predicting the next tick, but by identifying and following trends that last months or years.

The all-time high: the cleanest trend signal there is

How do you tell, objectively and without opinion, that a stock is in a healthy uptrend?

The simplest answer is also the most powerful: when it makes a new all-time high.

A stock at all-time highs has no one trapped at higher prices. There is no technical resistance overhead — because it has never been higher. There are no sellers waiting to "get back to even." It's the purest definition of strength.

Buying a new all-time high isn't buying high. It's buying confirmed strength.

What 33 years of the S&P 500 tell us

You don't have to take our word for it. We ran the numbers.

We took the full history of SPY (the S&P 500 ETF), from its January 1993 debut through 2026. We identified every single day it closed at a new all-time high — 636 in total — and measured where the index stood one year later (250 trading days) in each case.

The result:

• 86.6% of the time, the index was higher one year later.
• The average return across all 636 cases was +14.15%.
• When positive, the average gain was +17.78%; when negative (just 85 of 636 cases), the average loss was −9.46%.

Put plainly: historically, almost any day the S&P 500 broke to a new all-time high turned out to be a good entry point — provided you held for a year. You didn't have to pick the perfect day. You just had to buy strength and give it time.

The deeper pattern: the market breathes in cycles

There's something more important hiding in the data.

All-time highs don't spread evenly across time. They cluster. The S&P 500 printed dozens of new highs per year during 1995–2000, and again during 2013–2021 (79 in 2021 alone). Between 2001 and 2012 — the dot-com unwind and the 2008 crisis — it produced essentially zero new highs for years on end.

The market lives in two states: either it's trending, where highs beget highs, or it isn't, and they vanish for years.

And the few losing cases in our study weren't random. They concentrated at the end of each cluster — the highs made just before a major top (2000, late 2021). While a trend is alive, buying strength sits on the right side of probability. Danger only shows up when the trend exhausts itself.

From this, two rules govern how we operate:

1. While the trend lives, buy strength. A stock at new highs can keep printing new highs for months.
2. When the trend dies, step aside. And go elsewhere — at any given moment, hundreds of stocks are breaking out somewhere in the market.

For context: we're in one of those clusters right now. The S&P 500 printed fresh all-time highs this very week of May 2026, above $750, in the middle of an ongoing uptrend.

You won't always be right — and that's why risk matters

None of this means being right every time. Buy at the very top of a cycle and you lose. Our own study shows it: 13.4% of the time, you were down a year later.

But you don't need to be right every time. You need your wins to be larger than your losses. In the SPY study, average gains (+17.8%) nearly doubled average losses (−9.5%). That asymmetry is what keeps a system profitable over the long run, even when one in seven trades fails.

Stops, volatility, and the trap of chasing the top

There's another source of losses the SPY study doesn't capture — because it operates on an index, with no stop. On individual stocks, even within a healthy trend, volatility spikes can take you out at the stop before the move resumes. You accept it as part of the game: the goal isn't to capture the entire move, it's to capture the bulk of it. And it's certainly not to call the top of a trend — because the top is only visible after the fact.

Position-level risk control is the central piece of any breakout system. But that's a topic for another article.

It's not just the S&P 500: the evidence is broader

This isn't unique to the index, nor to our analysis alone.

In 1993, Narasimhan Jegadeesh and Sheridan Titman published the paper in The Journal of Finance that founded the academic momentum literature. They showed that buying past winners and selling past losers generated significant positive returns — returns that couldn't be explained by extra risk. The effect has since been replicated across markets worldwide, and the authors confirmed it again thirty years later.

Applied directly to buying highs: the study by Cole Wilcox and Eric Crittenden (Blackstar Funds, 2005) analyzed 18,000+ trades buying stocks at all-time highs and holding them until a volatility-based trailing stop was hit, across 22 years. Average return per trade: +15.2%, with a 49.3% win rate and a 2.56 win/loss ratio. At the portfolio level (1991–2008), the strategy returned 15.5% annually vs 7.9% for the S&P 500 — with a maximum drawdown of −29.3% vs −44.9%, less than half the index's. A 2024 replication confirmed the effect is still alive.

Three layers of evidence pointing at the same thing: intuition (the S&P 500), practice (Blackstar, on individual stocks with exit management), and academia (Jegadeesh-Titman, Nobel included).

From the idea to the tool

The conclusion is simple: in an asset that's trending, an all-time high is not a danger sign — it's participation in an uptrend already in motion. The hard part isn't the idea. It's executing it systematically.

Because if hundreds of stocks are breaking out at any given moment, you need a way to find them — all of them, every day, without bias and without opinion — among thousands of candidates.

That's why we built ATH Scanner: a U.S. market screener that systematically detects all-time-high breakouts across thousands of tickers. Each day, the filter narrows that universe down to around 300 stocks with strong Trend Return — the best of them in momentum — ranks them by the theoretical return of their trend (over 3, 6, or 12 months), and measures the cleanliness of that trend with a Quality (R²) score: a straight-line climb isn't the same as a chaotic chop to the same price. Sector and industry filters show where capital is flowing, and entry and stop prices come pre-calculated. A tool designed to find strength, not to opine about it. You decide what to trade.

---

Sources

• Fama, E., Hansen, L.P. & Shiller, R. — Nobel Prize in Economics 2013, "empirical analysis of asset prices."
• Jegadeesh, N. & Titman, S. (1993). Returns to Buying Winners and Selling Losers. The Journal of Finance, 48(1), 65–91.
• Wilcox, C. & Crittenden, E. (2005, updated 2009). Does Trend Following Work on Stocks? Blackstar Funds.
• Our own analysis: SPY all-time highs study, 1993–2026 — GitHub repository.